With the appropriate metabolic tracers positron emission tomography (PET) is able to measure tumor proliferation, potentially providing critical information about treatment response. While [C-11]thymidine can be employed in this task, its routine use is complicated by its short half-life and rapid in vivo catabolism. To overcome these difficulties this proposal will focus on [F-18]FLT (3'-deoxy-3'-fluorothymidine), which has been demonstrated to provide high quality images of tumors and proliferating tissues. FLT undergoes little degradation and employs F-18, a longer lived tracer. FLT is trapped in rapidly growing tumors by phosphorylation by thymidine kinase 1 (TK), an enzyme which increases about 10 fold as the cell enters the DNA synthetic phase. This proposal seeks to further simplify the use of FLT, better understand the variables that lead to its retention, and use it to measure response to therapy. The specific aims include: 1) To improve the use of FLT through a more rapid, automated synthesis. To assist in imaging tumors in the liver, where FLT is trapped by glucuronidation, further studies will be done to block uptake using probenecid. 2) Imaging with FLT of different tumor types as well as inflammatory lesions will be studied to explore the range of conditions where it will be useful. 3) To more fully understand the factors contributing to images of FLT, different quantitative approaches will be studied including kinetic modeling. These results will be compared to measurements of TK and other markers of tumor metabolism made on biopsy specimens. 4) Finally, the most important test of FLT will involve demonstrating its use as an early measure of tumor response to therapy in patients undergoing both standard cytotoxic therapy as well as new cytostatic treatments that are designed to delay tumor growth. These new cytostatic drugs, such as tyrosine kinase and angiogenesis inhibitors, an are difficult to assess using conventional anatomic imaging, since tumor shrinkage is not expected. Metabolic imaging may provide the only feasible way to determine if such agents are efficacious short of large phase III trials. It is particularly hard to determine the optimum dose and schedule of such drugs without an intermediate marker of response. These studies will test the hypothesis that PET imaging of proliferation will provide the best way to asses such agents. In summary, previous work in our laboratory has developed a novel and relatively straightforward way to image tumor proliferation. This proposal seeks to further understand the best way to quantitate, interpret, and clinically test this approach to measure treatment response.